{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YLJOXH7SKRIWLFBGKI5YBMCTY3","short_pith_number":"pith:YLJOXH7S","schema_version":"1.0","canonical_sha256":"c2d2eb9ff25451659426523b80b053c6e23c9d7d65693fc138ef928db04ecb5b","source":{"kind":"arxiv","id":"2603.29897","version":2},"attestation_state":"computed","paper":{"title":"UniRank: End-to-End Domain-Specific Reranking of Hybrid Text-Image Candidates","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Lei Xu, Lin Qu, Lin Yang, Shikui Tu, Wanxi Deng, Yupei Yang","submitted_at":"2026-02-08T12:39:19Z","abstract_excerpt":"Reranking is a critical component in many information retrieval pipelines. Despite remarkable progress in text-only settings, multimodal reranking remains challenging, particularly when the candidate set contains hybrid text and image items. A key difficulty is the modality gap: a text reranker is intrinsically closer to text candidates than to image candidates, leading to biased and suboptimal cross-modal ranking. Vision-language models (VLMs) mitigate this gap through strong cross-modal alignment and have recently been adopted to build multimodal rerankers. However, most VLM-based rerankers "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2603.29897","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-02-08T12:39:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"72afba2c0bfcb11721706b3d1eec5a7e2aa3d9c6ab5fe1264489ceb9c50ee35a","abstract_canon_sha256":"179a73c78b0872b72436c0e998fb0057c99fdb0925bd215b4d80f2a9fd15e87c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:29.105530Z","signature_b64":"TiK5wcjvV9c/jJlBaoyuzrYzhHadICcs6ggMNGVyoC+P7Z7rPLqY7pxhjN6a37Gkzjxhx7BswY4VCuGtVFjlDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c2d2eb9ff25451659426523b80b053c6e23c9d7d65693fc138ef928db04ecb5b","last_reissued_at":"2026-05-26T01:03:29.104598Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:29.104598Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UniRank: End-to-End Domain-Specific Reranking of Hybrid Text-Image Candidates","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Lei Xu, Lin Qu, Lin Yang, Shikui Tu, Wanxi Deng, Yupei Yang","submitted_at":"2026-02-08T12:39:19Z","abstract_excerpt":"Reranking is a critical component in many information retrieval pipelines. Despite remarkable progress in text-only settings, multimodal reranking remains challenging, particularly when the candidate set contains hybrid text and image items. A key difficulty is the modality gap: a text reranker is intrinsically closer to text candidates than to image candidates, leading to biased and suboptimal cross-modal ranking. Vision-language models (VLMs) mitigate this gap through strong cross-modal alignment and have recently been adopted to build multimodal rerankers. However, most VLM-based rerankers "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.29897","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2603.29897/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2603.29897","created_at":"2026-05-26T01:03:29.104759+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.29897v2","created_at":"2026-05-26T01:03:29.104759+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.29897","created_at":"2026-05-26T01:03:29.104759+00:00"},{"alias_kind":"pith_short_12","alias_value":"YLJOXH7SKRIW","created_at":"2026-05-26T01:03:29.104759+00:00"},{"alias_kind":"pith_short_16","alias_value":"YLJOXH7SKRIWLFBG","created_at":"2026-05-26T01:03:29.104759+00:00"},{"alias_kind":"pith_short_8","alias_value":"YLJOXH7S","created_at":"2026-05-26T01:03:29.104759+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YLJOXH7SKRIWLFBGKI5YBMCTY3","json":"https://pith.science/pith/YLJOXH7SKRIWLFBGKI5YBMCTY3.json","graph_json":"https://pith.science/api/pith-number/YLJOXH7SKRIWLFBGKI5YBMCTY3/graph.json","events_json":"https://pith.science/api/pith-number/YLJOXH7SKRIWLFBGKI5YBMCTY3/events.json","paper":"https://pith.science/paper/YLJOXH7S"},"agent_actions":{"view_html":"https://pith.science/pith/YLJOXH7SKRIWLFBGKI5YBMCTY3","download_json":"https://pith.science/pith/YLJOXH7SKRIWLFBGKI5YBMCTY3.json","view_paper":"https://pith.science/paper/YLJOXH7S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.29897&json=true","fetch_graph":"https://pith.science/api/pith-number/YLJOXH7SKRIWLFBGKI5YBMCTY3/graph.json","fetch_events":"https://pith.science/api/pith-number/YLJOXH7SKRIWLFBGKI5YBMCTY3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YLJOXH7SKRIWLFBGKI5YBMCTY3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YLJOXH7SKRIWLFBGKI5YBMCTY3/action/storage_attestation","attest_author":"https://pith.science/pith/YLJOXH7SKRIWLFBGKI5YBMCTY3/action/author_attestation","sign_citation":"https://pith.science/pith/YLJOXH7SKRIWLFBGKI5YBMCTY3/action/citation_signature","submit_replication":"https://pith.science/pith/YLJOXH7SKRIWLFBGKI5YBMCTY3/action/replication_record"}},"created_at":"2026-05-26T01:03:29.104759+00:00","updated_at":"2026-05-26T01:03:29.104759+00:00"}